Study On Radar Target Detection Method Based On Time Reversal

With the increasing application of radar in urban,hill,forest and other scenes,radar is also facing many new challenges,among which the multipath effect will seriously affect the detection performance of radar,resulting in unstable detection results or false targets.In order to avoid the influence of multipath on radar detection,traditional methods usually regard multipath signals as interference or clutter and suppress them.However,multipath signal contains target information.Hence,multipath exploitation is also one of the ideas to solve the multipath problem.Time reversal(TR)is realized by reversing the received signals in the time domain(conjugation in the frequency domain),normalizing the energy and then retransmitting to the scattering environment.The multipath channel is adaptive matched by the again received signals of radar,and the space-time focusing effect is achieved.Therefore,applying TR to radar target detection can achieve the effect of using multipath to improve the detection performance without knowing the multipath channel response in advance.The essence of TR is multipath channel adaptive matching.Therefore,the channel must be kept constant to obtain a complete matching effect.However,this condition is difficult to meet in many actual working scenarios of radars.For this problem,in the thesis,based on the TR technique,target detection of different radar systems,such as the single antenna,array antennas and multistatic radars,is studied for time-invariant and time-varying channels from simple to complex.The main content of this dissertation is summarized as follows.(1)The problem of stationary target detection based on TR technology in time-invariant channel is studied.In the time-invariant multipath channel scenario,the multipath effect will reduce the detection performance of the radar to the target.Aiming at this problem,a multistatic radar stationary target detection method based on TR is proposed,which comprehensively utilizes the multistatic radar’s spatial diversity characteristics and the TR technology’s multipath space-time focusing characteristics to improve the radar’s target detection performance.Combined with the time-invariant characteristics of the channel,considering the prior known and unknown multipath channels,a conventional likelihood ratio detector and a TR likelihood ratio detector without TR are designed.Through numerical simulation and theoretical analysis,it is proved that the detection performance of the multistatic radar detector using the TR technology is better than that of the conventional detector in the timeinvariant multipath environment.The detection effect of the target is better.Secondly,under the premise of ensuring that the total energy of the transmitted signal is consistent,according to the characteristics of multipath channel spectrum fluctuation,the TR technology is improved,and the echo is used to estimate the multipath channel and redistribute the energy of the TR transmitted signal.Based on this design,the frequency-optimized TR detector has a further improvement in detection performance compared to the TR detector.(2)The problem of moving target detection based on TR technology is studied.The essence of TR is channel adaptive matching,but in a multipath environment,target motion will cause channel changes,which makes the matching and focusing effect of TR technology on the channel worse.Aiming at the problem that the radar detection effect is poor in multipath channel,a conventional multipath exploitation method and a moving target detection method based on TR are proposed.Although the two transmission channels of TR have changed due to the influence of the target motion,on the premise that the channels are correlated and the statistical characteristics remain unchanged,the delay,amplitude,Doppler frequency and channel of the target echo to multipath are used.The parameters such as the variance of the change are estimated,and the channel change is incorporated into the TR detector,so that the detector not only realizes the matching and utilization of the multipath channel through the TR technology,but also adaptively perceives the channel change and improves the robustness of the detector.It solves the problem of poor detection performance of moving objects in a time-varying multipath environment.(3)The problem of target detection based on TR technology in time-varying channels is studied.In addition to target motion,factors such as changes in the background environment can also affect the channel response.In the scenario where a variety of factors comprehensively change the multipath channel,aiming at the problem of the degradation of focusing performance caused by the mismatch of the multipath channel during the application of the TR technology,a multistatic radar target detection method based on TR in the timevarying channel is firstly proposed.For the scene where the multipath channel obeys the complex Gaussian distribution and the statistical characteristics are constant,this method designs the TR detector from the perspective of the frequency domain.Limitations of applications in time-varying environments.Simulation and theoretical analysis prove that the detection method has strong robustness while using multipath to improve the detection performance.Secondly,a target detection method combining TR technology and polarization diversity characteristics is proposed for more general scenarios where the channel statistical characteristics change.This method relaxes the correlation restriction on the statistical characteristics of the time-varying channel,obtains the multipath channel response and channel disturbance response through maximum likelihood estimation,and designs a generalized likelihood ratio detector including channel disturbance.In addition,the performance gain of the polarized array antenna radar TR target detector compared with the conventional detector is quantitatively studied,which proves that the TR detector still has the detection advantage in the case of channel variation.